Employees' Retirement Plan of the National Education Association v. Clark County Education Association

UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA

EMPLOYEES’ RETIREMENT PLAN OF
THE NATIONAL EDUCATION
ASSOCIATION,

and

RETIREMENT BOARD OF THE
EMPLOYEES’ RETIREMENT PLAN OF
THE NATIONAL EDUCATION
Civil Action No. 20-3443 (RDM)
ASSOCIATION OF THE UNITED STATES,

Plaintiffs,

v.

CLARK COUNTY EDUCATION
ASSOCIATION,

Defendant.

MEMORANDUM OPINION

Plaintiff the Employees’ Retirement Plan of the National Education Association of the

United States (“the Plan”) is a multiemployer pension plan. Joint Appendix (“J.A.”) 4805.1

Defendant the Clark County Education Association (“CCEA”) is a labor organization that for

years was a contributing employer to the Plan. J.A. 4806. CCEA withdrew from the Plan in

2018, at which point the Plan assessed $3,246,349 in “withdrawal liability” against it. J.A. 188.

1
The Joint Appendix appears in seven parts at Dkt. 21. In addition to the Plan, the Retirement
Board of the Plan is also a Plaintiff in this action. According to the Complaint, the “Members of
the Retirement Board are fiduciaries within the meaning of ERISA Section 3(21)(A), . . . and
they bring this action in their fiduciary capacity on behalf of themselves and the . . . Plan’s
participants and beneficiaries for the purpose of collecting withdrawal liability and unpaid
contributions.” Dkt. 1 at 3 (Compl. ¶ 5).
CCEA challenged this assessment in arbitration and for the most part prevailed. The arbitrator

concluded, among other things, that the actuarial assumptions that the Plan used to calculate

CCEA’s withdrawal liability were unreasonable in the aggregate because one crucial assumption,

the discount rate (5.0%), was itself unreasonable. J.A. 6325–26 (Award at 2–3). So he ordered

the Plan to recalculate CCEA’s withdrawal liability using a different discount rate (7.3%). Id.

The Plan now asks this Court to vacate or modify most of that arbitration award, while

CCEA asks the Court to enforce it in full. The Court agrees with CCEA and the arbitrator that

the Plan’s assessment of CCEA’s withdrawal liability cannot stand, although it reaches that

result for different reasons than did the arbitrator. But, based on the arbitrator’s award, the Court

cannot determine whether the remedy the arbitrator imposed was permissible. It will therefore

remand the case to the arbitrator to reconsider the remedy. Because the Court denies the relief

that the Plan seeks but grants only a portion of the relief CCEA seeks, the Court will GRANT in

part and DENY in part the Plan’s motion for summary judgment, Dkt. 24, GRANT in part and

DENY in part CCEA’s cross-motion for summary judgment, Dkt. 26, and AFFIRM the

arbitration award in part and VACATE it in part.

I.

The Court begins by reviewing the relevant statutory, factual, and procedural

background.

A.

In a multiemployer pension plan, multiple employers make financial contributions to the

same general trust fund, and the money in that fund is used to provide for the pensions of the

various employers’ employees. 29 U.S.C. § 1002(37); see Concrete Pipe & Prods. Of Cal., Inc.

v. Constr. Laborers Pension Tr. for S. Cal., 508 U.S. 602, 605–06 (1993). These plans are

2
maintained in accordance with collective bargaining agreements between the employers and a

union and are governed by the provisions of ERISA. United Mine Workers of Am. 1974 Pension

Plan v. Energy West Mining Co., 39 F.4th 730, 734 (D.C. Cir. 2022). Among other things,

ERISA requires employers participating in multiemployer plans to “contribute annually to the

plan whatever is needed to ensure that it has enough assets to pay for the employees’ vested

pension benefits when they retire.” Id.

To estimate its annual funding needs, a plan makes assumptions about the relative rates at

which its assets and liabilities will grow. See Wachtell, Lipton, Rosen & Katz v. Comm’r, 26

F.3d 291, 293–94 (2d. Cir. 1994). A key assumption in this analysis is the “funding rate:” the

estimated annual rate of return the plan’s assets will earn. Chicago Truck Drivers Union v. CPC

Logistics, Inc., 698 F.3d 346, 353–54 (7th Cir. 2012). A higher funding rate represents a faster

assumed rate of growth for the plan’s assets and thus, all other things equal, necessitates lower

ongoing contributions from participating employers. Id. at 355. A lower funding rate,

conversely, represents a lower estimated growth rate of the plan’s assets and thus, all other things

equal, necessitates higher participant contributions. Id.

An employer who participates in a multiemployer plan is free to withdraw from the plan

and to terminate its obligation to make annual contributions. 29 U.S.C. § 1383. But an

employer’s withdrawal does not divest any worker enrolled in the plan of the pension benefits he

or she has earned; the plan and its remaining contributors must still provide for the vested

pension benefits of all its participants. See Energy West, 39 F.4th at 734–35 & n.2. This

structure can create perverse incentives: If a plan’s funding begins to lag—say, because a market

downturn decreases the value of its assets—participating employers will be required to make

larger annual contributions in order to comply with ERISA. Milwaukee Brewery Workers’

3
Pension Plan v. Jos. Schlitz Brewing Co., 513 U.S. 414, 416–17 (1995). And as required annual

contributions grow, so too does the incentive for participating employers to withdraw. Id.

Withdrawals further exacerbate funding shortfalls, and a shortfall-withdrawal-shortfall cascade

can send a plan into a “death spiral.” Energy West, 39 F.4th at 734. Although ERISA created a

federally chartered insurance corporation, the Pension Benefit Guaranty Corporation (“PBGC”),

to backstop troubled pension plans and to head off death spirals, 29 U.S.C. § 1302, in practice,

the existence of this safety net only further encouraged withdrawals and threatened to stretch the

PBGC’s obligations beyond its means, see Connolly v. Pension Benefit Guar. Corp., 475 U.S.

211, 214–15 (1986).

Congress enacted the Multiemployer Pension Plan Amendments Act of 1980 (the

“MPPAA”), Pub. L. 96-364, 94 Stat. 1208, to address this problem. To ensure that employers

pay their fair share (and to discourage strategic withdrawals), the MPPAA requires withdrawing

employers to pay for the privilege. 29 U.S.C. § 1381. Under the MPPAA, an employer that

withdraws from a multiemployer plan must pay “its pro rata share of the pension plan’s funding

shortfall,” also known as its withdrawal liability. CPC Logistics, 698 F.3d at 347; 29 U.S.C.

§ 1383(a), (b). More specifically, “withdrawal liability” is imposed based on “the employer’s

proportionate share of the plan’s ‘unfunded vested benefits,’ calculated as the difference between

the present value of the vested benefits and the current value of the plan’s assets.” Pension

Benefit Guar. Corp. v. R.A. Gray & Co., 467 U.S. 717, 725 (1984) (quoting 29 U.S.C. §§ 1381,

1391); see 29 U.S.C. § 1393(c).

Withdrawal liability is a function of both known variables and indeterminate

assumptions. For instance, when calculating withdrawal liability, a plan’s actuary knows how

many employees are enrolled in the plan and what benefits their pensions promise. But the

4
actuary must estimate, among other things, how long these employees will work and how long

they will live. Energy West, 39 F.4th at 735. The assumption with the greatest effect on the

withdrawal liability bottom line is the rate at which the plan’s assets will grow “by the miracle of

compound interest”—that is, the discount rate. CPC Logistics, 698 F.3d at 348. As in the

funding rate context, the higher the withdrawal liability discount rate, the faster the plan’s assets

are projected to grow on their own, and thus the smaller the present value of the plan’s liabilities,

the lower the funding shortfall, and the less a withdrawing employer’s withdrawal liability. See

id. And, conversely, the lower the discount rate, the slower the assets are assumed to grow, and

thus the greater the present value of the plan’s liabilities, and the more a withdrawing employer

must pony up. See id. The MPPAA requires plans calculating withdrawal liability to use

“actuarial assumptions and methods which, in the aggregate, are reasonable (taking into account

the experience of the plan and reasonable expectations) and which, in combination, offer the

actuary’s best estimate of anticipated experience under the plan.” 29 U.S.C. § 1393(a)(1).

Although withdrawal liability is paid over time, it is calculated only once, shortly after

the employer’s withdrawal. J.A. 6349 (Award at 26). This means that at the moment withdrawal

liability is assessed, risk—upside and downside—shifts from the withdrawing employer to the

plan and its remaining participants. See id. If the estimate of the withdrawing firm’s liability is

too low, the remaining employers will have to foot the bill for that shortfall and make sure that

the withdrawing firm’s employees (and other participants) still receive the pension benefits to

which they are entitled. See id. By the same token, if the withdrawal liability assessment is too

high, the plan can accrue the windfall, because the withdrawing employer has no way to recoup

its overpayment. See id.

5
 Withdrawal liability can be substantial, and, not surprisingly, plans and the employers

who withdraw from them often disagree about which assumptions to use. If a withdrawing

employer wants to dispute a plan’s calculations, it must do so through arbitration in the first

instance. 29 U.S.C. § 1401(a)(1). A plan’s determination of withdrawal liability receives

considerable deference in the arbitration process and is “presumed correct” by the arbitrator

unless the withdrawing employer “shows by a preponderance of evidence” that either the

actuarial “assumptions and methods” used were unreasonable “in the aggregate,” “taking into

account the experience of the plan and reasonable expectations,” or that the plan’s actuary “made

a significant error” in applying those assumptions or methods. Id. § 1401(a)(3)(B); see also id.

§ 1401(a)(3)(A) (noting more generally that “any determination made by a plan sponsor under

[the provisions governing calculation of withdrawal liability] is presumed correct unless the

party contesting the determination shows by a preponderance of the evidence that the

determination was unreasonable or clearly erroneous”). After arbitration, “any party can seek ‘to

enforce, vacate, or modify the arbitrator’s award’ in district court.” Energy West, 39 F.4th at 736

(quoting 29 U.S.C. § 1401(b)(2)).

B.

Here, the Plan is a multiemployer pension plan sponsored by the National Education

Association (“NEA”), a national labor union that represents teachers and other public-school

employees. J.A. 4805. CCEA became a contributing employer to the Plan in 2003 and served as

NEA’s local affiliate in the Clark County school district of Nevada. J.A. 4806. For the duration

of CCEA’s participation in the Plan, the parties’ relationship was governed by the “plan

documents,” a comprehensive set of rules setting forth the Plan’s structure, membership, policies

6
and procedures, the reciprocal rights and obligations of the Plan and its members, and the like.

J.A. 2–101.

Two series of events precipitated the instant lawsuit. The first was the Plan’s decision to

reassess and ultimately to revise its withdrawal liability discount rate assumption. Since 2014,

the Plan had used a 7.3% discount rate for calculating both annual funding obligations and

withdrawal liability. J.A. 569; J.A. at 2941–42 (Hudecek Dep.). In early 2017, however, the

Plan’s actuary began investigating how the Plan had determined its withdrawal liability

assumptions—the discount rate in particular—before he became the Plan’s actuary in 2015. J.A.

2946 (Hudecek Dep.). He raised the issue at a March 7, 2017 meeting of the Plan’s Board of

Directors, which appointed a committee to “investigate the withdrawal liability rate issue,”

presumably with an eye toward assessing whether the Plan’s assumption should be reconsidered.

J.A. 105. The committee made a presentation to the Board in May, and, at the September Board

meeting, the actuary recommended that the Plan lower the withdrawal liability rate—but not the

funding rate—to 5.0%, down from 7.3%. J.A. 116, 118, 121, 123. The Board agreed and

adopted that recommendation. J.A. 123.

According to the actuary, shifting the withdrawal liability rate from 7.3% to 5.0% was

justified because a rate of 5.0% reflected both a low-risk investment environment and the

expected returns on lower-risk, fixed-income investments. J.A. 3006 (Hudecek Dep.). He

explained that when an employer leaves the Plan, it is no longer “participating in any future risks

associated with the plan’s investments.” Id. As a result, “valuing a liability for that employer

based on a rate that provides for additional investment risk, which the other employers are

bearing after that employer’s withdrawal, did not make sense.” Id. Accordingly, he advised the

Board that “[a]n argument can be made that vested liabilities for withdrawing employers should

7
be determined based on interest rates representative of the low risk market environment,”

because a “low risk environment can potentially lessen the expected return on plan assets.” J.A.

133. After speaking with the Plan’s investment advisors, he determined that a discount rate of

5.0% “reflected a low-risk investment environment, particularly for fixed income,” J.A. 2969

(Hudecek Dep.), and would “eliminate not all, but much of [the] future investment risk [related

to the withdrawing] employer going forward,” J.A. 3006 (Hudecek Dep.); see also J.A. 2967,

2978, 3014, 3063–64 (Hudecek Dep.). In selecting the fixed-income assets on which to focus in

determining the discount rate, the actuary considered prior fixed-income returns for the Plan, the

Plan’s current fixed-income holdings, and the various fixed-income asset classes in which the

Plan was invested. J.A. 3015–16, 3021. But he did not aim to track or to predict the Plan’s

overall anticipated investment returns; rather, fixed-income investments represented only about

40% of the Plan’s holdings. J.A. 3021–22; see also J.A. 706 (noting in an actuarial valuation

report that the revised 5.0% withdrawal liability interest rate “was based on the actuary’s best

estimate of expected returns of low investment risk fixed income investments”).

Also in early 2017, a long-running dispute between CCEA and NEA’s Nevada state

affiliate, the Nevada State Education Association (“NSEA”), boiled over. See J.A. 3641 (Stocks

Dep.); J.A. at 5656–73 (Vellardita Test.). The particulars are not relevant here, but, in brief,

CCEA believed that it was not obtaining enough from NSEA in return for the amount of union

dues its members were paying. J.A. 3646 (Stocks Dep.); J.A. at 5657 (Vellardita Test.). After

efforts to negotiate a settlement failed, CCEA terminated its service agreement with NSEA,

began withholding dues, and eventually filed a lawsuit against NSEA. J.A. 3700, 3741, 3760

(Stocks Dep.); J.A. at 5667–69 (Vellardita Test.). On April 26, 2018, CCEA formally

disaffiliated from NEA and NSEA, stating that it would “no longer have any contractual

8
relationships” with either entity. J.A. 170. Several weeks later, the Plan terminated CCEA’s

membership and ended its participation. J.A. 182.

After terminating CCEA, the Plan assessed CCEA’s withdrawal liability at $3,246,349,

an amount reflecting the Plan’s recent shift to a 5.0% discount rate assumption for purposes of

assessing withdrawal liability. J.A. 188–89. ERISA provides that withdrawal liability attaches

when an employer “permanently ceases to have any obligation to contribute” under a plan, and in

the Plan’s view the termination of CCEA satisfied this criterion. Id. (quoting 29 U.S.C.

§ 1383(a)). CCEA requested review of the assessment, J.A. 225, and ultimately demanded

arbitration, J.A. 257–58.

C.

The parties submitted four issues to arbitration before the arbitrator, Adam Segal:

(1) Whether CCEA withdrew from the Plan during the Plan Year from
January 1, 2018 to January 31, 2018;

(2) Whether the actuarial assumptions and methods used by the Plan to
calculate[] [] CCEA’s withdrawal liability met the requirements of 29
U.S.C. § 1393(a)(1);

(3) Whether the Plan had and failed a legal obligation to timely and properly
provide [] CCEA with notice of the change to its actuarial assumption
for the discount rate of 5% utilized for calculation of withdrawal liability;
and

(4) Whether the Plan failed to comply with Section 12.4(b) of the Plan
document.

J.A. 6326 (Award at 3).

The arbitrator found for the Plan on the first issue, but CCEA ran the table on the others,

all of which pertained to whether the actuary’s discount rate assumption was permissible. J.A.

6332–33 (Award at 9–10). The arbitrator began his analysis of the second issue—whether the

actuarial assumptions and methods the Plan used to calculate CCEA’s withdrawal liability met

9
the requirements of 29 U.S.C. § 1393(a)(1)—by reviewing the relevant legal standards. He first

noted that the MPPAA requires that a plan’s actuary determine withdrawal liability using

“actuarial assumptions and methods which, in the aggregate, are reasonable (taking into account

the experience of the plan and reasonable expectations) and which, in combination, offer the

actuary’s best estimate of anticipated experience under the plan.” J.A. 6333 (Award at 10)

(quoting 29 U.S.C. § 1393(a)(1)). He then explained that to prevail on a challenge to such a

calculation under § 1401(a)(3)(B)’s aggregate unreasonableness threshold, a withdrawing

employer must show “that the combination of methods and assumptions employed in the

calculation would not have been acceptable to a reasonable actuary.” J.A. 6333–34 (Award at

10–11) (quoting Concrete Pipe, 508 U.S. at 635).

Applying these standards, the arbitrator concluded that a 5.0% discount rate did not

represent the actuary’s “best estimate” of anticipated experience under the Plan. J.A. 6350

(Award at 27); see 29 U.S.C. § 1393(a)(1). He accepted the proposition that, as a general matter,

the “use of a 5% [withdrawal liability] discount rate and a 7.3% funding rate is not prohibited.”

J.A. 6349 (Award at 26). This is because the two assumptions “serve different purposes under

the actuarial standards,” and the withdrawal liability rate appropriately may reflect the fact that

“a withdrawing employer no longer bears the investment risk the other employers bear.” J.A.

6349 (Award at 26). Nevertheless, he concluded that, in view of all of the circumstances, “a lack

of appropriate actuarial process” rendered the actuary’s 5.0% assumption neither reasonable nor

his “best estimate.” J.A. 3650–56 (Award at 27–33). These circumstances included: (1) the

magnitude of the difference between the funding and withdrawal liability rates, (2) the rapidity

of the Plan’s change of assumptions, and (3) the actuary’s failure to identify any factors that

changed so as to justify a new, materially lower rate. J.A. 6350 (Award at 27). This last

10
deficiency was the most glaring to the arbitrator: because the actuary had failed adequately to

identify why he changed this assumption, his actions did not conform to “accepted actuarial

practice.” J.A. 6350–53 (Award at 27–30). What the arbitrator considered to be the actuary’s

primary explanation for the shift—he had not focused on the withdrawal liability rate before,

because there had not been any withdrawals during his tenure as the Plan’s actuary—was, in the

arbitrator’s view, simply evidence of the actuary’s impermissible lack of diligence. J.A. 6353

(Award at 30).

The arbitrator next determined that 5.0% was not only an unreasonable assumption

standing alone, but one that rendered the actuary’s (and therefore the Plan’s) withdrawal liability

assumptions unreasonable in the aggregate under § 1401(a)(3)(B). J.A. 6356 (Award at 33).

The discount rate assumption was unreasonable on its own for the same reasons that it was not

the actuary’s “best estimate:” the actuary’s justifications for why he changed the rate from 5.0%

to 7.3% were, in the arbitrator’s view, inconsistent with reasonable actuarial practice. J.A. 6356–

57 (Award at 33–34). And because no other assumption offset the dramatic impact that the

discount rate change had on CCEA’s liability, the arbitrator concluded that the aggregate liability

calculation was unreasonable too. Id. He further reasoned that meeting ERISA’s aggregate

unreasonableness standard was not even necessary for CCEA to prevail in this case, because the

Plan had “elected to hold itself to an arguably higher standard for actuarial assumptions” in

section 12.4(b) of its plan documents, which requires the Plan to adopt a withdrawal liability

discount rate that is reasonable, without regard to whether the overall withdrawal liability

calculation is reasonable in the aggregate. J.A. 6356 (Award at 33).

That would have been enough to invalidate the Plan’s withdrawal liability assessment,

but the arbitrator also concluded that the Plan was required to provide CCEA with notice when it

11
changed the withdrawal liability discount rate but had failed to do so. J.A. 6357 (Award at 34).

Under 29 U.S.C. § 1394(b), plan sponsors must “give notice to all employers who have an

obligation to contribute under the plan . . . of any plan rules or amendments adopted pursuant to”

the MPPAA. The arbitrator recognized that, “[n]ormally,” an actuary’s change of the withdrawal

liability discount rate does not require notice, because it is neither a plan rule nor amendment to a

rule. J.A. 6357–58 (Award at 34–35). But “the unique facts” of the case convinced him that

notice was required here. J.A. 6358 (Award at 35). Because section 12.4 of the plan documents

makes the funding rate the default withdrawal liability rate absent an affirmative change by the

Board, the arbitrator considered it “misleading” for the Plan to make such a change without

notice. Id. The arbitrator also found that the Plan had failed to provide CCEA certain other

notices that were by all accounts statutorily required, which in his view “weigh[ed] heavily in

favor of finding” that a change to the funding rate was a plan rule subject to notice under

§ 1394(b). Id.

The arbitrator’s bottom line was this: because the Plan’s actuary had failed adequately to

justify the decision to lower the discount rate from 7.3% to 5.0%, he had to recalculate CCEA’s

withdrawal liability using a 7.3% discount rate—the Plan’s funding rate and previous withdrawal

liability rate—and assess CCEA’s liability accordingly. J.A. 6326 (Award at 3).

D.

The Plan filed this action on November 11, 2020, asking the Court to affirm the

arbitrator’s conclusion that CCEA had withdrawn from the Plan in 2018 but otherwise to vacate

and/or modify the arbitration award to uphold the Plan’s initial withdrawal liability assessment.

Dkt. 1 at 21 (Compl. ¶¶ 1–4). CCEA answered and counterclaimed, requesting the Court to

confirm and to enforce the arbitration award across the board. Dkt. 16 at 28. The parties agree

12
that no issues of material fact are in dispute, and, accordingly, they have cross-moved for

summary judgment. Dkt. 20 at 2; Dkt. 24; Dkt. 26.

II.

This case implicates two distinct standards of review: the typical summary judgment

standard under Federal Rule of Civil Procedure 56 and the specific standard by which the Court

reviews an ERISA arbitration award.

Starting with the more familiar of the two, summary judgment is warranted if a party can

“show[ ] that there is no genuine dispute as to any material fact and [that the party] is entitled to

judgment as a matter of law.” Fed. R. Civ. P. 56(a). “[I]n ruling on cross-motions for summary

judgment, the court shall grant summary judgment only if one of the moving parties is entitled to

judgment as a matter of law upon material facts that are not genuinely disputed.” See Muslim

Advocs. v. U.S. Dep’t of Just., 833 F. Supp. 2d 92, 98 (D.D.C. 2011).

A separate body of law governs the review of arbitration awards, and the parties spend

considerable time debating the appropriate standard of review in this case. Dkt. 24–1 at 17; Dkt.

26 at 32–34; Dkt, 28 at 20–22. It is settled law that when reviewing an arbitration award under

ERISA, the Court (1) presumes that the arbitrator’s findings of fact are correct “unless they are

rebutted ‘by a clear preponderance of the evidence’” and (2) reviews the arbitrator’s legal

conclusions de novo. Energy West, 39 F.4th at 737 (quoting 29 U.S.C. § 1401(c)). What the

phrase “a clear preponderance of the evidence” actually means is less obvious. What does it

mean, after all, for a fact “clearly” to be “more likely than not?” The few courts that have

wrestled with this issue have concluded, as the Supreme Court did under similar circumstances

in Concrete Pipe, that some “[r]epair” of the statutory test “is essential,” and, accordingly, have

read the phrase to establish a “clearly erroneous” standard of review for factual findings. Jos.

13
Schlitz Brewing Co. v. Milwaukee Brewery Workers’ Pension Plan, 3 F.3d 994, 998–99 (7th Cir.

1993), aff’d on other grounds, 513 U.S. 414 (1995); see also Manhattan Ford Lincoln, Inc. v.

UAW Local 259 Pension Fund, 331 F. Supp. 3d 365, 378–79 (D.N.J. 2018).

According to CCEA, however, determining the appropriate standard of review in this

case is yet more complicated, because, in its view, the Court must apply a similarly deferential

standard of review to mixed questions of law and fact. CCEA is correct that some courts have

held that that because “the dominant theme” in review of ERISA arbitrations “is deference,”

arbitrators’ decisions on mixed questions “may be set aside only for clear error.” Jos. Schlitz

Brewing Co., 3 F.3d at 999. But the D.C. Circuit has never said this and, instead, has observed

that the decisions of ERISA arbitrators are “fully reviewable to determine whether applicable

statutory law has been correctly applied.” Parmac, Inc. v. I.A.M. Nat’l Pension Fund Benefit

Plan A, 872 F.2d 1069, 1071 (D.C. Cir. 1989) (emphasis added); see also Combs v. Classic Coal

Corp., 931 F.2d 96, 102 (D.C. Cir. 1991) (stating that when reviewing an ERISA arbitration

award, “the district court ha[s] the duty of determining ‘whether applicable statutory law has

correctly been applied and whether the findings comport with the evidence’” (quoting I.A.M.

Nat’l Pension Fund Benefit Plan C v. Stockton TRI Indus., 727 F.2d 1204, 1207 n.7 (D.C. Cir.

1984))). Although far from definitive, this language at least suggests that the de novo standard

applies to the application of the law to the facts.

CCEA also argues that the Court should apply the even more deferential standard of

review from the Federal Arbitration Act (“FAA”), 9 U.S.C. § 1 et seq., to portions of the

arbitrator’s decision. It points out that although the MPPAA departs from the FAA’s standard of

review for statutory claims, it also establishes the FAA’s rules as the default absent a specific

departure. See Dkt. 32 at 6; 29 U.S.C. § 1401(b)(3). That being the case, CCEA urges the Court

14
to apply the FAA standard when reviewing those aspects of the arbitration award pertaining to

non-statutory claims, namely those that involve interpretations of the plan documents. Dkt. 32 at

6. The Plan, unsurprisingly, contends that the FAA’s standard of review should not apply at all.

See Dkt. 28 at 20–22.2

For present purposes, the Court need not resolve whether CCEA is correct about whether

either of these more deferential standards apply here. Following the parties’ initial round of

briefing in this case, the D.C. Circuit issued its decision in Energy West, 39 F.4th at 730, which

the Court discusses in detail below. Notably, CCEA now argues that the Court can decide this

case based solely on the rule announced in Energy West and that, if the Court agrees, it need not

address the alternative grounds set forth in the arbitrator’s decision. Dkt. 41 at 5 & n.1 (“In that

respect, Energy West substantially simplifies the disposition of this case.”). If the Court follows

this path, questions relating to the proper standard of review dissipate. The Court need not

decide, for example, whether the FAA standard of review governs the Court’s review of the

arbitrator’s interpretation of the Plan documents, which the arbitrator relied on only for purposes

of the alternative grounds for decision that CCEA now says the Court need not reach. And, to

the extent that the Court relies on the D.C. Circuit’s intervening decision in Energy West, as

CCEA proposes, the Court will not be called upon to decide whether the arbitrator correctly

applied the rule announced in that decision (long after the arbitrator issued his decision) to the

facts of this case.

2
Energy West, 39 F.4th at 737, put to rest CCEA’s contention that arbitration awards issued
under the MPPAA are to be reviewed under the FAA standard across the board. Dkt. 26 at 32.
That holding is consistent with the D.C. Circuit’s decision in I.A.M. Nat’l Pension Fund Benefit
Plan C v. Stockton Tri Indus., 727 F.2d 1204 (D.C. Cir. 1984), which observed that review of
arbitration awards under the MPPAA “is quite plainly to be distinguished” from the typical
review of arbitration awards, id. at 1207 n.7.
15
 Nor does the D.C. Circuit’s intervening clarification of the law require the Court to send

the matter back to the arbitrator to apply Energy West to the facts of this case in the first instance.

Neither party has asked the Court to do so, and, importantly, the facts necessary to apply Energy

West are undisputed. Just as an appellate court “can affirm a judgment on any basis adequately

preserved in the record below,” United States ex rel. Heath v. AT&T, Inc., 791 F.3d 112, 123

(D.C. Cir. 2015), a district court can uphold an ERISA arbitration award on alternative grounds

that, as here, follow unmistakably from an intervening and binding precedent.3

Consideration of the proper standard of review, as a result, brings things full circle,

taking the Court back to the traditional summary judgment standard. All agree that the Court

must review questions of law de novo. Since the relevant facts are undisputed, moreover, the

Court can apply that law to those facts. Because nothing more is required to decide this case, the

Court will leave the difficult questions CCEA poses regarding the standard of review for another

day.

III.

A.

Two statutory provisions govern an actuary’s determination of withdrawal liability and

the review of that determination. Their interpretation and the interplay between them resolves

this case.

3
As explained further below, infra at 32–37, the Court cannot ascertain whether it agrees or
disagrees with the arbitrator and CCEA regarding the proper remedy in this case, because the
arbitrator did not explain his choice of remedy. CCEA does not maintain, however, that the
relevant standard of review has any bearing on the question of remedy. Rather, it relies
principally on legal arguments and, to the extent it relies on the facts, it seems to suggest that the
Court can review the record itself. Dkt. 43 at 7–11.
16
 The first provision, 29 U.S.C. § 1393(a)(1), contains substantive standards that govern

actuarial estimates of withdrawal liability. It states:

Withdrawal liability . . . shall be determined by each plan on the basis of . . .
actuarial assumptions and methods which, in the aggregate, are reasonable
(taking into account the experience of the plan and reasonable expectations) and
which, in combination, offer the actuary’s best estimate of anticipated
experience under the plan[.]

Id. Section 1393(a)(1) thus imposes two limitations: (1) an actuary’s assumptions and methods

must be reasonable in the aggregate, “taking into account the experience of the plan and

reasonable expectations,” and (2) the methods and assumptions must in combination represent

the actuary’s “best estimate” of the plan’s “anticipated experience.” Id.

The second provision, 29 U.S.C. § 1401(a)(3)(B), is found in the MPPAA’s section on

“[r]esolution of disputes,” and it governs review of withdrawal liability determinations in

arbitration and litigation. Specifically, § 1401(a)(3)(B) sets forth the showing that an employer

must make to successfully challenge a Plan’s determination of withdrawal liability:

[T]he [plan’s] determination [of unfunded vested benefits] is presumed correct
unless a party contesting the determination shows by a preponderance of
evidence that

(i) the actuarial assumptions and methods used in the determination were, in
the aggregate, unreasonable (taking into account the experience of the plan
and reasonable expectations), or

(ii) the plan’s actuary made a significant error in applying the actuarial
assumptions or methods.

Id. § 1401(a)(3)(B). Subsection (a)(3)(B)(i) thus mirrors the aggregate reasonableness prong of

§ 1393(a)(1) and requires an employer to show by a preponderance of the evidence that a plan

failed to adhere to that requirement in order to prevail in a withdrawal liability challenge. Unlike

§ 1393(a)(1), however, it does not include a “best estimate” prong.

17
 In Concrete Pipe, the Supreme Court explained how § 1401(a)(3)(B) should be applied

when an employer challenges a plan’s withdrawal liability determination. The Court observed

that § 1401(a)(3)(B) targets the “aggregate reasonableness of the assumptions and methods

employed by the actuary,” not the specific result of any calculation. 508 U.S. at 635. Noting

that this reasonableness requirement must be understood in light of the objects to which it is

applied, the Court concluded that the reasonableness of actuarial assumptions and methods

should be judged “by reference to what the actuarial profession considers to be within the scope

of professional acceptability in making an unfunded liability calculation.” Id. In a dispute about

the calculation of withdrawal liability, an employer therefore bears the burden of showing “that

the combination of methods and assumptions employed in the calculation would not have been

acceptable to a reasonable actuary.” Id.; see also id. (noting that to prevail on such a challenge,

the employer must demonstrate that the actuary “has based a calculation on a combination of

methods and assumptions that falls outside the range of reasonable actuarial practice”). As the

Supreme Court recognized, this is a difficult burden for employers to meet, since there are “often

several equally correct” ways to approach “technical actuarial matters.” Id. (internal quotation

marks omitted).

The D.C. Circuit had articulated a similar rule two years earlier in Combs, which

involved a challenge to a plan’s withdrawal liability interest rate assumption. Combs affirmed a

district court’s vacatur of an arbitration award, where the arbitrator had concluded that the

withdrawal liability interest rate selected by the plan was unreasonably low. 931 F.2d at 98, 102.

Like the Supreme Court in Concrete Pipe, the Combs court emphasized that because “[g]reat

differences of opinion exist as to actuarial methods,” Congress “created the statutory

presumption in favor of withdrawal determinations expressly to forestall endless disputes ‘over

18
technical actuarial matters with respect to which there are often several equally correct

approaches.’” Id. at 99–100 (quoting S. 1076, The Multiemployer Pension Plan Amendments Act

of 1980: Summary and Analysis of Consideration, 98th Cong., 2d Sess. 21 (1980)). Under

§ 1401(a)(3)(B), therefore, actuaries have “wide latitude” to operate within a “range of

reasonableness.” Id. at 100.

As for § 1401(a)(3)(B), then, the guidance is clear: the withdrawal liability calculation

(more specifically, the calculation of unfunded vested benefits) of a plan’s actuary should be

upheld unless the employer can show that the actuary’s methods and assumptions “fall[] outside

the range of reasonable actuarial practice.” Concrete Pipe, 508 U.S. at 635. This understanding

tracks how the instant case was litigated before the arbitrator, with each side presenting expert

evidence about what constituted reasonable actuarial practices under the circumstances, and the

arbitrator applying actuarial standards to the facts of the case.

The role that § 1393(a)(1) plays in disputes regarding withdrawal liability is less obvious

at first glance. The text of the provision does not indicate whether it provides an independent

basis for setting aside a withdrawal liability assessment; by its terms, § 1393(a)(1) is directed

only at plans and their actuaries. See 29 U.S.C. § 1393(a)(1). In any event, the standalone force

of § 1393(a)(1)’s first, aggregate reasonableness prong is of little importance because

§ 1401(a)(3)(B) incorporates it essentially word-for-word. Section 1393(a)(1)’s second clause,

however—the “best estimate” prong—does not appear in § 1401(a)(3)(B), raising the question

whether that clause is enforceable in an arbitration and any ensuing litigation.

The D.C. Circuit addressed this question in Energy West, and it became one of several

courts in recent years to hold that the best-estimate requirement provides a basis for setting aside

a Plan’s withdrawal liability calculation and that it constrains actuarial discretion to a greater

19
degree than do general principles of actuarial practice. See Energy West, 39 F.4th at 738; Sofco

Erectors, Inc. v. Trs. of Ohio Operating Eng’rs Pension Fund, 15 F.4th 407, 421 (6th Cir. 2021);

GCIU-Emp. Retirement Fund v. MNG Enterprises, Inc., 51 F.4th 1092, 1099–1011 (9th Cir.

2022). The pension plan in Energy West set its withdrawal liability interest rate at about 2.75%,

an approximation of the return from a risk-free annuity. 39 F.4th at 736. The plan’s actuary

(like the actuary here) explained that this rate was appropriate because “when an employer

withdraws from a plan, it no longer bears any risk associated with that plan’s investment

performance,” so it should not benefit from an interest rate that reflects both the risk and return

associated with that performance. Id. The arbitrator upheld the assumption, placing “great

weight” on the Actuarial Standards of Practice, which endorsed the use of “a discount rate

implicit in annuity prices” for just this reason. Id. at 737. The district court affirmed the

arbitration award, following a line of out-of-circuit decisions holding that the best-estimate

requirement is a procedural requirement that the estimate be made by the plan’s actuary, rather

than anyone else, not a substantive constraint on the actuary’s choice of assumptions. United

Mine Workers of Am. 1974 Pension Plan v. Energy West Mining Co., 464 F. Supp. 3d 104, 116–

20 (D.D.C. 2020), overruled by 39 F.4th 730.

The D.C. Circuit reversed, holding that in order for an assumption to reflect the actuary’s

“best estimate of anticipated experience under the plan,” the assumption must be “based on the

plan’s particular characteristics.” Energy West, 39 F.4th at 738. Under this standard, the

discount rate assumption must estimate “how much interest the plan’s assets will earn based on

their anticipated rate of return.” 39 F.4th at 738; id. (stating that the discount rate must reflect

“the plan’s actual or anticipated investment experience”); id. at 740 (“While there may be a

reasonable range of estimates, the discount rate assumption cannot be divorced from the plan’s

20
anticipated investment returns.”). Applying that rule, the D.C. Circuit concluded that the use of a

risk-free rate “might be appropriate if a plan were invested in risk-free assets, or perhaps if it

planned to invest the withdrawal liability payments in risk-free assets.” Id. at 738. But, the court

reasoned, since the plan was “currently and project[ed] to be invested in riskier assets,” the

discount rate “must reflect that fact.” Id.

This rule, the court explained, follows from a plain reading of § 1393(a)(1), which ties

the best estimate to “anticipated experience under the plan.” See id. Of particular relevance

here, Energy West considered and rejected the plan’s argument that because the use of a risk-free

rate was consistent with accepted actuarial practice, it constituted an appropriate assumption,

remarking that “the MPPAA,” not an actuarial standard of practice, “is the law.” Id. at 740.

That is, to the extent that actuarial standards run contrary to the best reading of the statute, the

standards are unlawful and must give, “regardless of how widespread the unlawful practice is

among the profession.” Id. at 740 & n.8. The court also observed that because the discount rate

is the “most impactful” assumption in the withdrawal-liability calculation, “if the actuary selects

a discount rate that is not the ‘best estimate of anticipated experience under the plan,’ this error

will usually render the [entire withdrawal liability] calculation contrary to the MPPAA.” Id. at

739.

Although Energy West rejected the plan’s contention that the best-estimate requirement

lacks substantive import, it declined to embrace the employer’s sweeping contention that a plan’s

actuary must, invariably, use the same assumption for the funding rate and the withdrawal

liability rate. Id. at 741–43. That argument was premised in part on the textual similarities

between the statutory provisions governing minimum funding and withdrawal liability

assumptions. Opening Br. for Defendant-Appellant Energy West Mining Company at 2–3, 17–

21
20, United Mine Workers of Am. 1974 Pension Plan v. Energy West Mining Co., 39 F.4th 730

(D.C. Cir. 2022) (No. 20-7054), 2020 WL 7122123, at *2–3, 17–20. Just as the MPPAA

requires an actuary to calculate withdrawal liability using assumptions and methods that “in the

aggregate” are “reasonable” and that are his or her “best estimate of anticipated experience under

the plan,” 29 U.S.C. § 1393(a)(1), it also requires an actuary to calculate minimum funding

obligations using assumptions and methods “each of which is reasonable (taking into account the

experience of the plan and reasonable expectations)” and which “in combination, offer the

actuary’s best estimate of anticipated experience under the plan,” id. § 1084(c)(3). Such similar

statutory language, the employer maintained, should be read to mandate identical discount rate

assumptions. But, as the D.C. Circuit explained, although similar, the two provisions employ

“somewhat different statutory language.” Energy West, 39 F.4th at 742. Most notably, while the

withdrawal liability provision requires actuaries to “use assumptions ‘which, in the aggregate,

are reasonable,’” id. (emphasis added) (quoting 29 U.S.C. § 1393(a)(1)), the minimum funding

provision requires that actuaries “use assumptions ‘each of which is reasonable,’” id. (emphasis

added) (quoting 29 U.S.C. § 1084(c)(3)). Given the parallel language and the overlap in what

the actuary is measuring, the court said that the two assumptions “will invariably be similar” and

found it difficult to imagine circumstances that would justify a large difference in the discount

rates (say, five hundred basis points). Id. But, the court affirmed, “it does not follow that the

discount rates must be identical.” Id.

Energy West further held that although only § 1401(a)(3)(B) is expressly directed at the

review of actuarial assumptions, a violation of § 1393(a)(1) also provides grounds for rejecting a

plan’s determination of withdrawal liability and for setting aside an arbitration award. See id. at

740–41. In Energy West, as here, the plan had argued that a failure to observe the best-estimate

22
requirement did not justify vacating an arbitration award, because the MPPAA’s dispute

resolution provision—§ 1401(a)(3)(B)—“does not specify that [§ 1401(a)(3)(B)’s] presumption

of correctness can be overcome by showing that the assumptions were not the best estimate of

anticipated experience under the plan.” Id. at 740 (internal quotation marks omitted). But the

D.C. Circuit was unpersuaded, explaining that for the same reasons that the actuary’s

assumptions were not his “best estimate of anticipated experience under the plan,” they were also

unreasonable “taking into account the experience of the plan and reasonable expectations.” Id.

Despite this analytical shift from § 1393(a)(1) to § 1401(a)(3)(B), the court made clear that it

was not announcing a § 1401(a)(3)(B) reasonableness holding with dicta about the best-estimate

requirement. Id. at 741. To the contrary, it emphasized that the best-estimate requirement was

the crux of the matter. Id. (“The arbitration award must be vacated because in determining the

withdrawal liability for Energy West, the actuary failed to use a discount rate that reflected the

Plan’s characteristics and was the ‘best estimate of anticipated experience under the plan.’”).

One way or another, under Energy West a withdrawal liability estimate at odds with a plan’s

actual investments cannot be upheld.4

4
The Court notes that, fairly read, “anticipated experience under the plan” could conceivably
encompass more than “predicted performance of the plan’s asset portfolio.” In Combs, for
example, the court suggested that a plan might reasonably have selected a conservative interest
rate assumption—even one that “did not account for the higher rate of return actually
experienced by the [p]lan[]”—because of the plan’s “weak funding position” and “the generally
unstable condition of the coal industry.” 931 F.2d at 100. More generally, a plan’s “anticipated
experience” might be understood to incorporate the plan’s withdrawal expectations, need to
insulate its participating employers from asymmetric downside risk without setting onerous
annual contribution requirements, the way in which the industries in which its participants
operate might impact their future financial viability, and the like. Energy West did not say that
such considerations are off limits, nor did it purport to undercut Combs, but it did make clear that
the actuarial assumptions must reflect the actuary’s best estimate of the anticipated experience
taking into consideration the plan’s actual investments and its projected investment returns.
23
 How then to reconcile Energy West and Concrete Pipe? Recall that Concrete Pipe said

that as far as § 1401(a)(3)(B) is concerned, standards of actuarial practice are the touchstone of

reasonableness. See 508 U.S. at 635. If such standards countenance a wide variety of methods

and assumptions, so too does the statute. See id. And because § 1401(a)(3)(B) sets the bar for

displacing a plan’s determination of withdrawal liability, Concrete Pipe could be read to hold

that so long as a plan’s methods and assumptions comport with industry standards, the plan is in

the clear. Energy West, on the other hand, treats § 1393(a)(1) as a substantive constraint on how

a plan must assess withdrawal liability, one that stands apart from and above industry practice.

39 F.4th at 740. A plan’s failure to abide by that § 1393(a)(1) constraint, moreover, can render

the actuary’s estimate unreasonable for purposes of § 1401(a)(3)(B). Id. at 741. Insofar as

Energy West reads § 1393(a)(1) to constrict the “range of reasonableness” under § 1401(a)(3)(B)

more tightly than does standard actuarial practice, it arguably precludes a plan from relying on

actuarial methods and assumptions that Concrete Pipe would seemingly permit.

The Sixth Circuit’s recent decision in Sofco Erectors, Inc. v. Trustees of Ohio Operating

Engineers Pension Fund, 15 F.4th 407 (6th Cir. 2021), offers one way to reconcile such a

restrictive best-estimate requirement with Concrete Pipe. It reads Concrete Pipe to address only

the question of actuarial discretion as to how to estimate something, not what an actuary is

supposed to estimate. Id. at 423. That is, § 1393(a)(1) tells actuaries what to estimate:

“anticipated experience under the plan.” According to Sofco Erectors (and Energy West), that

means an estimate that reflects the plan’s “actual portfolio,” and actuarial standards have no

bearing on that “policy choice,” which Congress made when it enacted the MPPAA. 15 F.4th at

421, 423. But, so long as the actuary is estimating the actual “anticipated experience under the

24
plan,” he or she has discretion under Concrete Pipe to choose among various permissible

actuarial practices to do so. Id. at 423.

At any rate, Energy West is both on-point and binding, and it establishes the following

approach for adjudicating a best-estimate challenge to a withdrawal liability discount rate

assumption: A court (or arbitrator) must first determine, with appropriate deference to the plan,

whether the rate that the plan selected “reflect[s] the [p]lan’s characteristics,” namely, “the

[p]lan’s past or projected investment returns.” Energy West, 39 F.4th at 740, 741. If so, the

court then looks to actuarial practice and Concrete Pipe to assess whether the assumptions and

methods used to make that estimate were consonant with reasonable actuarial practice. See

Sofco Erectors, 15 F.4th at 423. But if not, the assumption does not reflect the actuary’s “best

estimate of anticipated experience under the plan” and is therefore unreasonable. 29 U.S.C.

§ 1393(a)(1). The court must then determine whether that unreasonable assumption renders the

plan’s actuarial assumptions and methods unreasonable “in the aggregate,” “taking into account

the experience of the plan and reasonable expectations.” Id. § 1401(a)(3)(B)(i). In cases

involving a material variation in the discount rate, the answer will usually be yes, and if it is, the

Plan’s withdrawal liability assessment cannot stand. This analysis would of course generally be

conducted by an arbitrator in the first instance, but the arbitration in this case occurred before

Energy West clarified the law, and neither the Plan nor CCEA has asked the Court to remand the

case to the arbitrator to apply Energy West in the first instance.

B.

1.

It is evident from the record that the 5.0% withdrawal liability discount rate that the

Plan’s actuary selected was not his “best estimate of anticipated experience under the plan” as

25
Energy West interpreted that language. The thrust of the actuary’s explanation to the Board for

his 5.0% recommendation was that 5.0% reflected the return of “low risk” fixed-income

investments. J.A. 133; see J.A. at 706 (noting in the 2018 actuarial valuation that the Plan’s

change in the withdrawal liability discount rate was based on “the actuary’s best estimate of

expected returns of low investment risk fixed[-]income investments”). The problem is that the

anticipated returns of low-risk fixed-income investments in the abstract lack the required

connection to the Plan’s “particular characteristics” and its portfolio’s expected returns. See

Energy West, 39 F.4th at 738. The actuary’s deposition testimony demonstrates as much and

makes clear that his choice of discount rate was not tied to his professional assessment of the

likely return on the Plan’s actual assets. Critically, he acknowledged that a majority of the Plan’s

assets were not invested in fixed income, estimating that only about 40% of the assets were. J.A.

3021–22 (Hudecek Dep.) (acknowledging that “around 40 percent” of the fund’s investments

were in fixed income securities); J.A. 3574 (Kra Expert Report) (noting that “[t]he actual

portfolio of the Fund is approximately 30% bonds and 70% equities/real estate/other.”). 40%

does not cut it. Energy West requires that an actuary “estimate how much interest the plan’s

assets will earn based on their anticipated rate of return.” 39 F.4th at 738. A discount rate

assumption based on the expected returns on a type of asset that makes up less than half of a

Plan’s portfolio falls short of that standard.

To be sure, the 5.0% discount rate assumption was not entirely detached from the

actuary’s views about the Plan’s portfolio. At deposition, he testified that the expected fixed-

income returns that formed the basis of his estimate corresponded to the fixed-income assets held

by the Plan. J.A. 3015–16 (Hudecek Dep.) (noting that he “look[ed] at what fixed income

returns were in prior years for the [P]lan” and had “conversations with investment managers in

26
regard to estimated future experience in regard to those fixed incomes that are currently being—

were invested in the [P]lan”). He also testified that he at least looked at “all . . . asset[] classes

that were particular under the plan” and that he “didn’t focus on asset classes that the plan had no

assets in.” J.A. 3021 (Hudecek Dep.). And he told the Board that his estimate reflected a “low

risk market environment,” J.A. 133, and the prevailing market environment is no doubt an

important determinant of the Plan’s “anticipated experience.” But none of this can overcome the

fact that, by the actuary’s own admission, his assumption ultimately failed to take account of the

majority of the Plan’s investments.

Energy West does not deprive actuaries of all flexibility, and, consistent with the statute

and long-settled precedent, the decision recognizes that there can be a “range” of permissible

discount-rate assumptions. 39 F.4th at 742. Nor does Energy West hold that an actuary’s

estimate must encompass the expected return of all of the plan’s assets. But it does preclude

estimates that self-consciously disregard the expected returns of the majority of these assets. An

estimate is not “based on the Pension Plan’s past or projected investments returns” if it fails to

take account of likely returns on most of the fund’s investments. Id. at 740.

The actuary’s deposition testimony also clarifies that he focused on a low-risk asset class

that did not comprise the bulk of the Plan’s portfolio because of his views about risk shifting. He

explained that when an employer withdraws from a plan, that employer no longer bears any of

the risk “associated with the plan’s investments.” J.A. 3006. In his view, it “did not make

sense” to ignore that fact in setting a withdrawal liability discount rate and, essentially, to

compensate the employer for risk it was no longer bearing. Id. The statement in his declaration

that “[t]he 5% rate reflected [his] best estimate of an anticipated risk-adjusted rate in light of the

27
experience of the Plan and its expectations” reflects that theory of withdrawal liability. J.A.

2927 (Hudecek Decl. ¶ 19) (emphasis added).

Energy West does not preclude all consideration of risk shifting. It permits it, however,

only to the extent that such consideration is part of the actuary’s assessment of expected

investment returns; doing so cannot justify ignoring expected returns. Energy West recognized

that because the ultimate statutory standard is one of reasonableness, the best-estimate

requirement “does not mandate adopting any single numerical assumption,” as long as an

actuary’s assumptions are within an “acceptable range” and “based on the plan’s actual

characteristics.” 39 F.4th at 742. Accounting for shifting risk is a way of introducing a measure

of conservatism into a withdrawal liability estimate, and Energy West quoted with approval the

Second Circuit’s admonition that the statute “is not violated when an actuary chooses an

assumption that is within the range of reasonable assumptions, even when the assumption is at

the conservative end of the range.” Wachtell, Lipton, Rosen & Katz, 26 F.3d at 296. The fact

that a plan’s remaining participants must absorb the asset risk previously associated with a

withdrawing member would also seem to be one of the plan’s “characteristics” and part of its

“anticipated experience.” 29 U.S.C. § 1393(a)(1). So the Court does not understand Energy

West to bar actuaries from weighing risk shifting in the course of selecting an assumption “at the

conservative end” of a range of reasonable estimates of a plan’s “anticipated investment returns.”

39 F.4th at 739–40. But an actuary cannot risk shift his way to a discount rate “divorced from” a

plan’s anticipated returns, id. at 740 or, as in this case, the majority of the assets that drive such

returns.

28
 2.

After the D.C. Circuit issued its decision in Energy West, this Court granted the parties

leave to submit supplemental briefs addressing the decision. Min. Order (Nov. 7, 2022); Dkt. 40,

Dkt. 41; Dkt. 42; Dkt. 43. Unsurprisingly, CCEA argues that Energy West “substantially

simplifies the disposition of this case” and that it “plainly and emphatically holds that” a

discount rate untethered to actual plan experience “is not lawful.” Dkt. 41 at 5, 13. The Plan, by

contrast, insists that the discount rate that its actuary selected was permissible under Energy

West. It points to four main pieces of evidence in support of this contention: First, the arbitrator

found that the Plan’s actuary “reviewed historical, current, and projected economic information

derived from his conversations with the Plan’s investment advisor and a review of fixed income

returns” and “considered all factors contained in section 3.8 of ASOP 27, looking into those he

considered relevant to the determination.” Dkt. 40 at 7 (quoting J.A. 6345 (Award at 22)). Such

express consideration of the Plan’s characteristics distinguishes this case from Energy West, says

the Plan, because the actuary in Energy West did not seem to have considered that plan’s

characteristics at all. Id. Second, the arbitrator concluded that “a 5% discount rate, in isolation

from the other issues in this case, is within the range of rates selected by reasonable actuaries in

accordance with actuarial standards” and that the use of a 5% discount rate for withdrawal

liability and a 7.3% rate for funding purposes is not prohibited. J.A. 6349 (Award at 26); see

Dkt. 40 at 8. Third, the arbitrator also, at least in principle, approved the actuary’s methodology

of “using a set rate based on low-risk bond rates, one that is somewhat close to the result yielded

by common blended rates.” J.A. 6350 (Award at 27). Finally, the Plan’s expert opined that

5.0% is a reasonable estimate of the expected return of the Plan’s entire portfolio, inclusive of all

its assets, not just its fixed-income holdings. Dkt. 40 at 9 (citing J.A. 3574–75 (Kra Expert

29
Report)). All of this, according to the Plan, is powerful evidence that 5% falls “within an

acceptable range” “based on the actual characteristics of the plan,” id. at 6 (quoting Energy West,

39 F.4th at 742), which is all that Energy West requires, id. at 5–9.

For the most part, the Court has little quarrel with these contentions. The undisputed

evidence shows that the Plan’s actuary “looked at” all of the Fund’s “asset class[es],” J.A. 3021

(Hudecek Dep.), and the arbitrator did not find otherwise, J.A. 6350–55 (Award at 27–32).

Similarly, no one has asked the Court to overturn the arbitrator’s general observations about a

5.0% discount rate and the actuary’s basic methodology. Nor, finally, does the Court see any

reason not to credit the conclusion of the Plan’s expert that 5.0% could be a reasonable estimate

of the expected returns of the Plan’s entire portfolio.

But none of this changes the fact that the Plan’s actuary acknowledged that he only

accounted for the anticipated experience of a category of assets comprising approximately 40%

of the Plan’s investments when setting the discount rate. Energy West does not allow such an

approach. The fact that the actuary considered a portion of the Plan’s assets cannot make up for

the fact that he did not consider most of them. If the actuary had taken a more comprehensive

look at the Plan’s portfolio, it is possible that he, like the Plan’s expert, might still have

concluded that 5.0% was the appropriate discount rate. But the Court is concerned only with

what the actuary actually did, not what he might have done. CCEA’s withdrawal liability, after

all, must be based on “the actuary’s best estimate of anticipated experience under the plan,” not

someone else’s best estimate of anticipated experience under the plan or the actuary’s best

estimate of something else, even if that something else might conceivably align with a

reasonable estimate of anticipated experience under the plan. 29 U.S.C. § 1393(a)(1).

30
 C.

Under the logic of Energy West, the Plan’s discount rate assumption rendered its

withdrawal liability assumptions unreasonable in the aggregate. Although Energy West did not

equate § 1393(a)(1)’s best-estimate requirement with § 1401(a)(3)(B)’s presumption that a plan’s

determination of withdrawal liability is correct unless an employer can show by a preponderance

of the evidence that the assumptions and methods used were “in the aggregate, unreasonable

(taking into account the experience of the plan and reasonable expectations),” it noted that the

two provisions substantially overlap. 39 F.4th at 741. As the court explained, both provisions

require consideration of a plan’s “experience.” Id. As a result, just as failing to consider a plan’s

characteristics offends the best-estimate requirement, “[i]f the actuary is not basing the

assumptions on the plan’s characteristics, the assumptions will not be reasonable ‘taking into

account the experience of the plan.’” Id. (quoting 29 U.S.C. § 1401(a)(3)(B)). Even before

Energy West, moreover, it was well-established that “an erroneously low interest rate

assumption” can “destroy the validity of the entire [withdrawal-liability] calculation,” unless that

“apparently unreasonable assumption is offset by other assumptions, so as to produce an

assumption package that is reasonable in the aggregate.” Combs, 931 F.2d at 101.

Here, the discount rate assumption was unreasonable because it did not give due regard to

the Plan’s experience. The Court also finds no error in the arbitrator’s conclusion that the Plan’s

overall calculation contained no “offsetting changes to blunt its impact.” J.A. 6357 (Award at

34). So the Plan’s calculation of CCEA’s withdrawal liability was unreasonable “in the

aggregate.” 29 U.S.C. § 1401(a)(3)(B)(i). Resisting this conclusion, the Plan half-heartedly

gestures at the determination of its expert that the actuary’s administrative expense estimate was

31
“potentially” too low. Dkt. 24-1 at 18 n.4. Even if true, that falls well short of salvaging the

Plan’s assumptions as a whole.

IV.

Because Energy West and the undisputed evidence establish that the Plan’s actuarial

assumptions were unreasonable in the aggregate, the Plan’s withdrawal liability assessment must

be set aside. On this much, the Court agrees with the arbitrator’s bottom line, albeit on

alternative grounds. But the arbitrator took the further step of ordering the Plan on remand to

recalculate CCEA’s withdrawal liability using the 7.3% discount rate “applicable prior to the

change to 5%.” J.A. 6326 (Award at 3). He did not, however, provide an explanation for

imposing this remedy. Under these circumstances, the Court concludes that the best course is to

remand the case to the arbitrator to reconsider the question of remedy consistent with the

following discussion.

The Court notes at the outset that when an arbitrator concludes that a plan’s withdrawal

liability calculation must be set aside, it will often be appropriate to send the issue back to the

plan’s actuary to recalculate withdrawal liability, rectifying whatever defect the arbitrator

identified. As Congress recognized in enacting the MPPAA, calculating withdrawal liability

requires the exercise of actuarial judgment, and actuaries have discretion in formulating

reasonable assumptions that, collectively, offer their “best estimate of anticipated experience

under the plan.” 29 U.S.C. § 1393(a)(1); see also Concrete Pipe, 508 U.S. at 635; Combs, 931

F.2d at 102 (“Great differences of opinion exist as to actuarial methods. Congress, therefore,

created the statutory presumption in favor of withdrawal determinations . . . .”); Bd. of Trs.,

Mich. United Food and Com. Workers Unions v. Eberhard Foods, Inc., 831 F.2d 1258, 1261

(6th Cir. 1987) (“The statute does not anticipate that the actuary will always choose the figure

32
the court would choose as the most reasonable from among th[e] range [of reasonable

assumptions]. Rather, the only requirement is that in every case the actuarial determination will

fall within the range of reasonableness—taking into account the unique characteristics of the

fund.”). Settled precedent, moreover, recognizes that there is an “acceptable range” for actuarial

assumptions and that, although one would expect the discount rates used to calculate withdrawal

liability and minimum funding requirements to be similar, they need not be identical. Energy

West, 39 F.4th at 742 (internal quotation marks omitted).

Thus, if an actuary simply fails to offer a sufficiently detailed explanation for her

decision or applies a rule that the courts subsequently clarify, it will likely make sense to provide

the actuary another chance to explain herself or to re-run her calculations. In this sense, the

process of reviewing arbitration awards under ERISA is not so different from judicial review of

the actions of administrative agencies. Stockton TRI Indus., 727 F.2d at 1207 n.7. In both

circumstances, Congress has vested discretion in experts and instructed those conducting review

of those experts’ decisions—both courts and arbitrators—not to substitute their judgment for that

of those better versed in technical matters. Compare 29 U.S.C. § 1401(a)(3), with Motor

Vehicles Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983).

In some cases, however, it may be appropriate for an arbitrator to be more prescriptive.

In New York Times Co. v. Newspaper and Mail Delivers’—Publishers’ Pension Fund, for

example, the court imposed a discount rate after concluding that the rate the plan’s actuary

imposed was impermissible, because the actuary testified that the plan’s funding rate was in fact

her “best estimate of how the Pension Fund’s assets . . . will on average perform over the long

term.” 303 F. Supp. 3d 236, 255 (S.D.N.Y. 2018). The New York Times arbitrator reasonably

could have done the same in the first instance. See also Sofco Erectors, Inc. v. Trs. of Ohio,

33
Operating Eng’rs Pension Fund, No. 2:19-cv-2238, 2020 WL 2541970, at *10 (S.D. Ohio May

19, 2020), vacated in part on other grounds by 15 F.4th 407 (6th Cir. 2021) (imposing a discount

rate); GCIU-Emp. Retirement Fund v. MNG Enter., Inc., No. 21-61, 2021 WL 3260079, at *5

(C.D. Cal. July 8, 2021), vacated in part and remanded on unrelated grounds by 51 F.4th 1092

(same). Similarly, if a plan’s actuary has repeatedly failed to select a lawful rate, a specific

remedy may be appropriate. Cf. Pub. Citizen Health Rsrch. Grp v. Comm’r, Food & Drug

Admin., 740 F.2d 21, 32 (D.C. Cir. 1984) (“When agency recalcitrance is in the face of a clear

statutory duty or is of such magnitude that it amounts to an abdication of statutory responsibility,

the court has the power to order the agency to act to carry out its substantive statutory

mandates.”).

In short, ERISA arbitrators must strike a balance. In general, they must defer to

reasonable assumptions made by plan actuaries and must avoid substituting their own views for

those of the actuaries. But that deference has limits. And in appropriate cases, arbitrators have

the flexibility to impose different remedies in different circumstances. That flexibility is

reflected in the American Arbitration Association’s Multiemployer Pension Plan Arbitration

Rules (“AAA Rules”), which apply here, and which authorize an arbitrator to “grant any remedy

or relief within the scope of ERISA,” American Arb. Ass’n, Multiemployer Pension Plan

Arbitration Rules for Withdrawal Liability Disputes 13 (2020), https://www.adr.org/sites/

default/files/Multiemployer_Pension_Plan_Withdrawal_Liability.pdf.

The question whether the arbitrator in this case acted within his authority when he

directed the Plan to recalculate CCEA’s withdrawal liability using the 7.3% discount rate is

complicated by the fact that the arbitrator did not explain why that remedy—as opposed to a

34
more open-ended remedy—was appropriate. His decision, moreover, can be read in different

ways. One reading supports the remedy he imposed, and the other does not.

First, the actuary might have believed that, as a general matter, an arbitrator should

mandate the use of a discount rate if he determines that the discount rate the plan’s actuary

selected cannot stand. As explained above, if that is what he thought, he misapprehended the

respective roles of actuaries and arbitrators in the statutory scheme. In particular, one of the

arbitrator’s primary reasons for disapproving the discount rate the actuary selected was that the

actuary’s explanation for selecting that rate was inadequate. J.A. 6350–55 (Award at 27–32).

Where an actuary simply fails to provide a sufficient explanation for his decisions, a remand will

typically be preferable to specific relief.

But there is another possibility. Portions of the arbitrator’s decision at least suggest that

he made a factual finding that 7.3% was in fact the actuary’s best estimate of the appropriate

discount rate. Although the award never says this directly, it suggests as much. For example,

the arbitrator noted that the actuary had not identified what variables had changed so as to cause

him to reevaluate his prior discount rate assumption, stating that “[t]he evidentiary record . . .

leaves the strong impression that nothing was mentioned because nothing had truly changed.”

J.A. 6354 (Award at 31). And he considered it significant that the actuary’s 2018 valuation

stated that “[a]ll” of the assumptions involved, presumably including the new 5% discount rate,

were “based on the results of an experience study completed in 2015.” Id. This was, recall, the

same study that the actuary had relied on to set the withdrawal liability discount rate at 7.3% in

years prior. Id. And in the actuary’s view it was “simply inexplicable” that the “same

procedures and the same original study [yielded] a materially different result.” J.A. at 6355

(Award at 32). Thus the arbitrator might have reasoned that because the actuary had not stated

35
which, if any, of the underlying theoretical and empirical inputs that determine the withdrawal

liability discount rate had changed—and in fact there were some indications that none had

changed—7.3% remained the actuary’s best estimate. See id. If such a factual finding was the

basis for the remedial decision, the Court would owe deference to it. Factual findings are

reviewed under the clear error standard, Jos. Schlitz Brewing Co., 3 F.3d at 998–99, and the

Court cannot say that it would have been clearly erroneous to find on this record that 7.3% was,

in fact, the actuary’s actual best estimate of anticipated experience under the plan. If that is the

case, requiring the Plan to use that rate would be appropriate.

Ultimately, though, the arbitrator’s reasoning is unclear. Contrary to what CCEA claims,

the arbitrator did not make “a specific factual finding as to the Plan Actuary’s actual best

estimate as of the time in question,” Dkt. 43 at 7, and he did not connect any such finding to the

remedy he fashioned. But, on the other hand, portions of his decision gesture in that direction.

Without an explanation of the basis for the remedy, however, the Court cannot determine

whether that remedy should be enforced. The Court will therefore remand the matter to the

arbitrator to reconsider the issue. If, in light of the Court’s opinion, the arbitrator concludes that

giving the actuary another go is more appropriate than imposing a discount rate, he should do so.

But if he concludes that, despite the actuary’s testimony to the contrary, the actuary had really

believed that a discount rate of 7.3% best reflected the Plan’s anticipated experience, he should

say so.

* * *

The Court need not reach any of the other aspects of the arbitration award. The parties no

longer dispute that CCEA withdrew from the Plan, triggering withdrawal liability. Dkt. 1 at 14

(Comp. ¶ 60); see Dkt. 16 at 27 (Counterclaim ¶ 44). More to the point, CCEA has invited the

36
Court to disregard the other issues and to resolve the case based on Energy West alone, see Dkt.

41 at 5, and reaching the alternative grounds in the arbitration award would do nothing to benefit

the Plan. Finally, a remand is necessary in any event because the arbitrator did not explain what

bearing his various alternative holdings had on the remedy he imposed.

CONCLUSION

For the foregoing reasons, the Court will GRANT in part and DENY in part Plaintiffs’

motion for summary judgment, Dkt. 24, and GRANT in part and DENY in part Defendant’s

cross-motion for summary judgment, Dkt. 26. The Court will AFFIRM and ENFORCE the

arbitration award, except that it will VACATE the arbitrator’s order that “[t]he withdrawal

liability assessment in this case must be recalculated using the 7.3% [withdrawal liability]

discount rate applicable prior to the change to 5%.” The Court will also REMAND the case to

the arbitrator to reconsider the appropriate remedy in a manner consistent with this opinion.

Finally, the Court will STAY CCEA’s obligation to make further withdrawal liability payments

pending further order of the arbitrator or this Court.

A separate order will issue.

/s/ Randolph D. Moss
RANDOLPH D. MOSS
United States District Judge

Date: February 27, 2023

37